ACCURACY OF ORIGINAL GAS IN PLACE ESTIMATION BY MATERIAL BALANCE IN RADIAL WATER-DRIVE DRY-GAS RESERVOIRS

Abstract

One of the commonly used methods to estimate the original gas in place (OGIP) for water-drive dry-gas reservoirs is the material balance method, the (GpBg+WpBw)/(Bg-Bgi) versus We/(Bg-Bgi) plot (water-drive form) or the p/z versus Gp plot (depletion-drive form). Due to tank assumption of material balance, this method can lead to some errors of the estimation under certain circumstances. In order to investigate the error of estimated OGIP from these plots for cases having different permeability, aquifer sizes, shut-in durations, and amounts of historical data, a hypothetical dry-gas reservoir was created in a reservoir simulator. Production data from the simulation were used to make plots in order to determine OGIPs using four different water influx models, namely, simple aquifer model, Fetkovich, van Everdingen & Hurst, and Carter & Tracy. Results from this study show that p/z versus Gp plot with enough amount of historical data is still applicable if aquifer size is not larger than 10 PV. The (GpBg+WpBw)/(Bg-Bgi) versus We/(Bg-Bgi) plot with proper shut-in durations and water influx models is applicable for aquifer size up to 100 PV. The error increases when the aquifer size increases but decreases when the permeability increases. When shut-in duration increases, the errors in 50 mD cases decrease but the errors in 500 mD cases are not affected. If the aquifer size is unknown, OGIP can still be accurately estimated if the actual aquifer size is not larger than 30 PV in 50 mD cases and 100 PV in 500 mD cases.